Sex-specific functional effect of IL-12 gene polymorphisms in brain tumours

Abstract

The role of the IL-12 family cytokines in brain tumourigenesis is an active area of research, but only a few studies have explored the impact of the functional IL12B polymorphisms on brain tumour (BT) risk. The present study aimed to evaluate the possible impact and functional effect of IL12B rs3212227 and rs17860508 on BT susceptibility in adults. We observed a lower frequency of the C-allele variant of the rs3212227 polymorphism among cases with primary BT compared to those with brain metastasis (BrM) and individuals with no benign or malignant tumours under allelic and dominant genetic models. The frequency of the C-allele of rs3212227 was significantly lower in primary BT cases than in controls (16% vs. 26.5% with OR = 0.529, 95%CI: 0.3–0.94, p = .029; OR_adjusted=0.442; 95%CI: 0.23–0.87; p = .018). There was no significant association between rs17860508 and BT risk under all analysed genetic models after age and sex adjustment. A significantly higher IL-12p40 serum level in men with primary BT than in women (p = .0007) was found. We observed a sex-specific effect of rs3212227 polymorphism on serum IL-12p40 levels in cases with primary BT in contrast to BrM. The C-allele carriership was associated with higher IL-12p40 in women with primary BT. Serum levels of IL-23 were similar across the patients’ subgroups. The obtained results suggest that the IL12B rs3212227 polymorphism influences the risk for primary BT in contrast to rs17860508 polymorphism and shows a sex-specific functional effect on IL-12p40 levels. Also, the male gender was associated with higher IL-12p40 in cases with primary BT.

Keywords:

• IL-12p40
• IL-23
• primary brain tumours
• metastatic brain tumours
• rs3212227
• rs17860508

Introduction

Primary brain tumours (BT) are a heterogeneous group of tumours arising from cells within the central nervous system (CNS), among which gliomas represent a major type of malignancy in adults. In addition, brain metastases (BrM) are the most common intracranial tumour in adults, exceeding by ∼ five-fold primary BT [1]. Little is known about the immune mechanisms in BrM and primary BT, but several studies have suggested a different immune landscape. A review of literature [1] has outlined that, in contrast to glioblastoma multiforme (GBM), BrM shows moderate or even pronounced T cell infiltration, suggesting that the pattern of T cell infiltration depends on the primary tumours that metastasize to the brain [2,3]. The different rates of T cell exhaustion may explain the poor response to immune checkpoint blockade in GBM, in contrast to the efficacy of checkpoint inhibitors usage in patients with melanoma brain metastasis [4]. The BrM and primary BT differ in their number of tumour-infiltrating lymphocytes, microenvironment and tumour mutational burden [5]. However, further researches are needed to fully understand the fundamental difference between the genetic and immune background of primary and secondary BT.

Cytokines are important regulatory molecules in the innate and adaptive immune response. The interleukin 12 (IL-12) family of cytokines includes heterodimeric IL-12, interleukin-23 (IL-23), interleukin-27 (IL-27), interleukin-35 (IL-35) and recently discovered interleukin-39 (IL-39). They are composed of several subunits, such as p40, p35, p19, p28 and Epstein-Barr virus-induced gene 3 (EBI3). The p40 subunit is shared by IL-12, also known as IL-12p70 (IL12p35/p40) and IL-23 (IL-23p19/p40). The IL-12 family cytokines are involved in regulating the immune response, particularly the T-cell-mediated immune response. IL-12 and IL-23 promote Th1 and Th17 responses, respectively, while IL-35 has been shown to suppress Th1 and Th17 responses and promote the differentiation of regulatory T cells (Tregs). In addition, IL-12p40 is induced in excess over the other specific subunits of p19 and p35 and is known to exist in monomeric or homodimeric forms. Early studies in mouse models and human airway epithelial cells suggested a functional activity of IL-12p40/p80, as a macrophage chemoattractant and a competitive antagonist of IL-12p70 and IL-23 [6,7]. Overall, cytokines of the IL-12 family play a critical role in regulating the immune response and are involved in the pathogenesis of several diseases such as cancers and autoimmune disorders. Logically, IL-12 is a target in some ongoing clinical trials for anti-cancer therapy in human glioma [8–10]. The usage of the fused antibody-cytokine (immunocytokines) as IL-2, IL-12 or TNF-α was offered as a new strategy and showed promising results in mouse glioma models [11].

Human IL-12p40 is encoded by the IL12B gene located on chromosome 5q31-33. The rs3212227 polymorphism in the IL12B gene is a single nucleotide polymorphism (SNP), located in the 3’untraslated region (3’UTR) of a gene that has been associated with alterations in the production of IL-12p40 in cancer and autoimmune disorders [12,13]. Previous studies have shown that the rs3212227 polymorphism may be associated with an increased risk of developing cancer, and the variant rs3212227*C allele may act as a risk factor [14,15]. The genotype carriers of variant C-allele were associated with a significantly increased risk of glioma [16]. Another study in the Chinese population failed to find any association between the IL12B rs3212227 polymorphism and the risk of glioma [17]. A functional genetic variant in the IL12B gene was reported in a promoter polymorphism with rs17860508 which is due to a 4-bp micro-insertion (TTAGAG allele) combined with an AA/GC transition (GC allele). Several studies reported that rs17860508 alters gene expression and cytokines production [13,18,19]. However, to the best of our knowledge, the association between rs17860508 and the risk of primary and metastatic brain tumours has not been investigated yet.

Taking into consideration the role of the IL-12 family cytokines in brain tumourigenesis, and the limited number of studies on the impact of the functional IL12B polymorphisms on BT risk, the present study aimed to evaluate the possible impact and functional effect of rs3212227 and rs17860508 on BT susceptibility in adults.

Subjects and methods

Subjects

A total of 120 cases with primary or metastatic BT were included in the present study. They were recruited in the Neurosurgical Clinic at the University Hospital for Active Treatment ‘Prof Dr Stoayn Kirkovich’, Stara Zagora, Bulgaria. Demographic and clinical data, including gender, age of diagnosis, tumour type and location, date of surgical resection and accompanying diseases were collected from each patient. The group of patients consisted of 65 males (54.2%) and 55 females (45.8%) at ages ranging from 28 to 82 years (mean age 62.1 ± 10.5 years). The group of primary BT (n = 50) included eight cases with benign tumours - meningothelial meningioma (n = 6) and epidermoid cysts (n = 2), four cases with low-grade astrocytomas I–II grade (2 cases with grade I and 2 cases with grade II); 3 cases with anaplastic astrocytoma (grade III) and 35 cases with GBM (grade IV). Seventy (58.8%) patients were with BrM of different primary origin. Most BrM originated from lung cancer (n = 36; 51%) followed by colon or rectal cancer (n = 9; 13%); kidney or bladder (n = 5; 7%); melanoma (n = 4; 6%); breast (n = 3; 4%); endometrium (n = 2; 3%) and other or of unknown origin (n = 11). The histopathological types of brain tumours are presented in Table 1.

Table 1. Histopathological type of tumour.

Histopathological type of tumour	Number of cases
Primary brain tumour	50
Meningioma	6
Epidermoid cysts	2
Astrocytomas grade I	2
Astrocytomas grade II	2
Astrocytomas grade III	3
Glioblastoma multiforme (grade IV)	35
Brain metastasis	70
Lung cancer	36
Colon or rectal	9
Kidney or bladders	5
Melanoma	4
Breast	3
Endometrium	2
Other/unknown	11

The control group consisted of ethnically matched, unrelated individuals who did not have any benign or malignant tumours. The associative study of the impact of IL12B rs3212227 and rs17860508 polymorphism was conducted using a group of 240 controls (96 male and 144 female; mean age 39.4 ± 14.4 years), and a group of 331 controls (119 male and 212 female; mean age 39.8 ± 12.7 years), respectively. A smaller group (n = 74) of age and sex-matched healthy individuals was used as a control group for the cytokines quantification analysis. All included cases and controls were of Caucasian origin. The demographic and clinical parameters of the cases and controls included in this study are summarized in Table 2.

Table 2. Demographic and clinical parameters of the studied groups.

Characteristics	Brain tumours		Healthy volunteers	p Value
Number	120		240/331^a
Gender
Male, n (%)	65 (54%)		96 (40%) 119 (36%)	χ^2=6.495; p = .011; χ^2=12.098; p = .001
Female, n (%)	55 (46%)		144 (60%) 212 (64%)
Age
Mean ± SD, year	62.1 ± 10.55	39.4 ± 14.4/39.8 ± 12.7^a		p < .001 (t-test)
Min, year	28	24
Max, year	82	84
Tumour types
Low-grade glioma (WHO I–II), n (%)	4 (3%)
High-grade glioma (WHO III–IV), n (%)	38 (32%)
Meningioma, n (%)	6 (5%)
Other, n (%)	2 (2%)
Metastasis, n (%)	70 (58%)
Hemispheric location
Left, n (%)	49 (41%)
Right, n (%)	42 (35%)
Bihemispheric, n (%)	9 (7%)
Not applicable, n (%)	20 (17%)

^aControl group for IL12B rs3212227 polymorphism (n = 240) and for IL12B rs17860508 (n = 331); GBM: Glioblastoma Multiforme.

Genotyping

Genomic DNA was extracted from venous blood samples using a GeneMATRIX Basic DNA Purification Kit (cat. number E3545-01; EURx Ltd, Poland) following the manufacturer’s instructions. The concentrations and purity of the obtained DNA samples were determined spectrophotometrically using a BioDrop µLite+ (cat. number 80-3006-55 Biochrom Ltd., UKs).

Genotyping of rs17860508 and rs3212227 in the IL12B gene was previously described in detail [20]. Briefly, the genotyping of rs17860508 was performed by an amplification refractory mutation system (ARMS), also termed allele-specific polymerase chain reaction (PCR). The TTAGAG allele was marked as allele-1, while the GC allele as allele-2. The sequence of the primers was as follows: CTCTAA allele-1: 5’TGTCTCCGAGAGAGGCTCTAA-3′; GC allele − 2: 5′-TGTCTCCGAGAGAGGGCTGT-3′ and the common primer 5′-TGGAGGAAGTGGTTCTCGTAC3’. The amplification conditions were as follows: initial denaturing at 95 °C for 15 min, followed by 30 cycles: denaturation at 95 °C, annealing at 65 °C and final extension at 72 °C. Genotyping of rs3212227 was performed by restriction–fragment length polymorphism–PCR. The sequence of the primers was as follows 5′-ATTTGGAGGAAAAGTGGAAGA-3′ and 5′-AATTTCATGTCCTTAGCC ATA-3′. The amplification conditions were as follows: initial denaturing at 95 °C for 5 min, followed by 30 cycles: denaturation at 95 °C, annealing at 54.3 °C and final extension at 72 °C. The C-allele variant induces a restriction site of TaqI enzyme (cat number R0149S; New England Biolabs, Ipswich, MA, USA), which results in two fragments: 906 bp and 140 bp. The PCR reagents and primers were supplied by EURx Ltd, Poland and Metabion International AG, Germany (cat number cat. no. E2500; cat. number E0503), respectively. The PCR reactions were performed in a GeneAmp PCR System 9700 (Applied Biosystems, USA). The PCR products or restriction fragments were visualized by agarose gel electrophoresis, stained with ethidium bromide and detected by the Herolab gel documentation system and its E.A.S.Y (cat. number 28 08 010; Herolab GmbH Laborgeräte, Germany).

Quantification of cytokine serum levels

Quantification of cytokine serum levels was performed by enzyme-linked immunosorbent assay (ELISA) following the manufacturer’s instructions (Elabscience, USA). The detection range of the IL-12p40 ELISA kits (cat. number E-EL-H0151) was 31.25–2000 pg/mL and its sensitivity 18.75 pg/mL. The detection range of IL-23 ELISA kits (cat. number E-EL-H0107) was 39.06–2500 pg/mL and sensitivity − 23.44 pg/mL. The determination of the optical density was measured by an ELISA plate reader Thermo Scientific Multiskan EX (cat. number 51118170; Thermo Fisher Scientific Inc). The results were calculated according to the standard curves, constructed by using the kit’s provided standards, following the manufacturer’s instructions, and are expressed as picograms per millilitre (pg/mL).

Statistical analyses

The data were analysed using Statistica version 12 (StatSoft Inc., Tulsa, OK, USA). Conditional logistic regression with 95% confidence intervals (CIs) was utilized to identify the associations between IL12B polymorphisms and BT risk. A multivariate linear regression model adjusted for age and sex was also applied. The chi-square analysis was used to compare gender distribution, and the independent sample t-test to compare age between the groups. Variables with non-normal distribution, evaluated by the Shapiro-Wilk test, were analysed by non-parametrical analysis. Due to the non-normally distributed data of the cytokine serum levels, the Mann-Whitney U test was used. The data are reported and presented as mean values with standard deviation (±SD) and standard error (±SE) aimed to be useful for further large-scale meta-analysis. Differences were considered statistically significant at the level of p < .05.

Results

Clinical characteristics of the study participants

The demographic and clinical characteristics of the cases and controls included in the study are presented in Table 2.

In our study, there was higher frequency of BT in men in comparison to women (54% vs. 46%). There was also a higher prevalence of metastatic BT among men than women (61% vs. 39%). Women were predominantly diagnosed with primary BT (56%) mainly with meningioma, and lower-grade gliomas. Cases with GBM were in approximately equal percentages between both genders. The age at diagnosis in the cases with primary and metastatic brain tumours was similar (p = .123) and significantly higher than that of controls (p < .001).

Association of IL12B gene polymorphisms rs3212227 and rs17860508 with BT risk

The genotypes and allele frequencies of IL12B rs3212227 and rs17860508 polymorphisms among the controls and cases are summarized in Table 2.

The results from the general genetic models, including the codominant, the dominant, the recessive model, the overdominant models and the allelic model are presented in Table 3. The genotype distributions of both studied polymorphisms among the cases and the controls followed the Hardy-Weinberg equilibrium (chi-squared < 2.2; p > .3).

Table 3. Genetic models of the IL12B rs3212227 and rs17860508 polymorphisms among the cases with brain tumours and controls.

Rs3212227	Genotype/ variant	Cases, n (%) N = 120	Controls, n (%) N = 240	OR (95% CI); p value	Adjusted OR (95% CI); p value
Codominant model	AA	73 (61%)	134 (55.8%)	1.00	1.00
	AC	39 (32%)	85 (35.4%)	0.842 (0.52–1.35); p = .48	0.974 (0.53–1.8); p = .93
	CC	8 (7%)	21 (8.8%)	0.699 (0.3–1.66); p = .42	0.687 (0.23–2.04); p = .5
Dominant model	AA	73 (61%)	134 (55.8%)	1.00	1.00
	AC + CC	47 (39%)	106 (44.2%)	0.814 (0.52–1.27); p = .37	0.913 (0.51–1.62); p = .76
Recessive model	AA + AC	112 (93%)	219 (91.2%)	1.00	1.00
	CC	8 (7%)	21 (8.8%)	0.745 (0.32–1.74); p = .5	0.693 (0.24–2.01); p = .5
Overdominant model	AA + CC	81 (67.5%)	155 (64.6%)	1.00	1.00
	AC	39 (32.5%)	85 (35.4%)	0.878 (0.55–1.4); p = .58	1.017 (0.56–1.85); p = .96
Allelic model	A-allele	185 (77%)	353 (73.5%)	1.00	1.00
	C-allele	55 (23%)	127 (26.5%)	0.878 (0.63–1.22); p = .44	1.017 (0.67–1.56); p = .94
rs17860508	Genotype/ variant	Cases, n (%) N = 120	Controls, n (%) N = 331	OR (95% CI); p value	Adjusted OR (95% CI); p value
Codominant model	11	23 (19%)	97 (29%)	1.00	1.00
	12	52 (43%)	162 (49%)	1.354 (0.78–2.35); p = .28	1.004 (0.50–2.01); p = .99
	22	45 (38%)	72 (22%)	2.636 (1.47–4.74); p = .001	1.401 (0.66–2.98); p = .38
Dominant model	11	23 (19%)	97 (27%)	1.00	1.00
	12 + 22	97 (81%)	234 (71%)	1.748 (1.05–2.92); p = .033	1.154 (0.6–2.21); p = .67
Recessive model	11 + 12	75 (62.5%)	259 (78%)	1.00	1.00
	22	45 (37.5%)	72 (22%)	2.158 (1.37–3.39); p = .001	1.39 (0.77–2.52); p = .27
Overdominant model	11 + 22	68 (57%)	169 (51%)	1.00	1.00
	12	52 (43%)	162 (49%)	0.798 (0.52–1.22); p = .292	0.810 (0.47–1.39); p = .45
Allelic model	Allele-1	98 (41%)	356 (54%)	1.00	1.00
	Allele-2	142 (59%)	306 (46%)	1.686 (1.25–2.27); p = .001	1.214 (0.83–1.79); p = .33

Notes: BrM: brain metastasis; BT: brain tumours; OR: odds ration; 95% CI: 95% confidential interval; adjusted OR: adjusted odds ratio for age and sex; allele-1: TTAGAG allele of rs17860508; allele-2: GC allele of rs17860508. The significant associations are bolded.

As shown in Table 3, the observed genotypic and allelic frequencies of the IL12B rs3212227 polymorphism between cases and controls were similar. The rs3212227 polymorphism was not significantly associated with BT risk in all analysed genetic models.

The frequency of the rs17860508 allele-2 was significantly higher in the BT cases than in the controls, which suggested that allele-2 could be associated with a higher risk of BT (59% vs. 46% with OR = 1.686, 95% CI: 1.25–2.27, p = .001). Genotype-22 in the codominant and recessive models, as well as the presence of allele-2 (genotypes 12 + 22) in the dominant model, were associated with a higher risk of BT. However, the observed significant associations were lost after adjustment for demographic factors – age and sex. Our results confirmed that age is an important concomitant for the risk of BT (OR = 1.015, 95% CI: 1.0–1.03, p = .009).

Next, we analysed the distribution of rs3212227 and rs17860508 IL12B polymorphisms among cases with BT after stratification to primary and secondary BT (Table 4).

Table 4. Association between rs3212227 and rs17860508 IL12B polymorphisms and risk of primary BT or BrM in inheritance models.

Rs3212227	Genotype/ variant	Primary BT, n (%) N = 50	Controls, n (%) N = 240	OR (95% CI); p value	Adjusted OR (95% CI); p value	BrM, n (%) N = 70	Controls, n (%) N = 240	OR (95% CI); p value	Adjusted OR (95% CI); p value
Codominant model	AA	35 (70%)	134 (55.8%)	1.00	1.00	38 (54%)	134 (55.8%)	1.00	1.00
	AC	14 (28%)	85 (35.4%)	0.631 (0.32–1.24); p = .182	0.72 (0.34–1.55); p = .402	25 (36%)	85 (35.4%)	1.037 (0.59–1.84); p = .901	1.249 (0.62–2.54); p = .538
	CC	1 (2%)	21 (8.8%)	0.182 (0.02–1.4); p = .102	0.174 (0.02–1.48); p = .110	7 (10%)	21 (8.8%)	1.175 (0.46–2.97); p = .733	1.248 (0.39–4.04); p = .712
Dominant model	AA	35 (70%)	134 (55.8%)	1.00	1.00	38 (54%)	134 (55.8%)	1.00	1.00
	AC + CC	15 (30%)	106 (44.2%)	0.542 (0.28–1.04); p = .067	0.445 (0.2–0.99); p = .048	32 (46%)	106 (44.2%)	1.065 (0.62–1.82); p = .819	1.054 (0.65–2.43); p = .503
Recessive model	AA + AC	49 (96%)	219 (91.2%)	1.00	1.00	63 (90%)	219 (91.2%)	1.00	1.00
	CC	1 (4%)	21 (8.8%)	0.213 (0.03–1.62); p = .135	0.193 (0.023–1.63); p = .131	7 (10%)	21 (8.8%)	1.159 (0.47–2.85); p = .748	1.146 (0.37–3.6); p = .815
Overdominant model	AA + CC	36 (72%)	155 (64.6%)	1.00	1.00	45 (64%)	155 (64.6%)	1.00	1.00
	AC	14 (28%)	85 (35.4%)	0.709 (0.36–1.4); p = .316	0.814 (0.38–1.74); p = .596	25 (36%)	85 (35.4%)	1.013 (0.58–1.77); p = .963	1.209 (0.61–2.41); p = .59
Allelic model	A-allele	84 (84%)	353 (73.5%)	1.00	1.00	101 (72%)	353 (73.5%)	1.00	1.00
	C-allele	16 (16%)	127 (26.5%)	0.529 (0.3–0.94); p = .029	0.442 (0.23–0.87); p = .018	39 (28%)	127 (26.5%)	1.073 (0.7–1.64); p = .742	1.005 (0.976–1.036); p = .731
rs17860508	Genotype/ variant	Primary BT, n (%) N = 50	Controls, n (%) N = 331	OR (95% CI); p value	Adjusted OR (95% CI); p value	BrM, n (%) N = 70	Controls, n (%) N = 331	OR (95% CI); p value	Adjusted OR (95% CI); p value
Codominant model	11	8 (16%)	97 (29%)	1.00	1.00	15 (21%)	97 (29%)	1.00	1.00
	12	23 (46%)	162 (49%)	1.721 (0.74–4.0); p = .207	1.301 (0.52–3.27); p = .576	29 (41%)	162 (49%)	1.158 (0.59–2.27); p = .67	0.819 (0.36–1.86); p = .819
	22	19 (38%)	72 (22%)	3.2 (1.33–7.72); p = .01	1.8 (0.68–1.4.79); p = .240	26 (37%)	72 (22%)	2.335 (1.15–4.73); p = .018	1.154 (0.48–2.78); p = .479
Dominant model	11	8 (16%)	97 (27%)	1.00	1.00	15 (21%)	97 (27%)	1.00	1.00
	12 + 22	42 (84%)	234 (71%)	2.176 (0.985–4.8); p = .054	1.482 (0.62–3.53); p = .374	55 (79%)	234 (71%)	1.52 (0.82–2.82); p = .184	0.948 (0.44–2.03); p = .891
Recessive model	11 + 12	31 (62%)	259 (78%)	1.00	1.00	44 (63%)	259 (78%)	1.00	1.00
	22	19 (38%)	72 (22%)	2.205 (1.18–4.13); p = .014	1.486 (0.72–3.05); p = .280	26 (37%)	72 (22%)	2.126 (1.23–3.69); p = .007	1.321 (0.67–2.62); p = .426
Overdominant model	11 + 22	27 (54%)	169 (51%)	1.00	1.00	41 (59%)	169 (51%)	1.00	1.00
	12	23 (46%)	162 (49%)	0.889 (0.49–1.61); p = .698	0.891 (0.46–1.74); p = .736	29 (41%)	162 (49%)	0.738 (0.44–1.24); p = .254	0.746 (0.39–1.42); p = .371
Allelic model	Allele-1	39 (39%)	356 (54%)	1.00	1.00	59 (42%)	356 (54%)	1.00	1.00
	Allele-2	61 (61%)	306 (46%)	1.82 (1.18–2.8); p = .006	1.486 (0.72–3.05); p = .28	81 (58%)	306 (46%)	1.597 (1.11–2.31); p = .013	1.321 (0.67–2.62); p = .426

Notes: BrM: brain metastasis; BT: brain tumours; OR: odds ration; 95% CI: 95% confidential interval; adjusted OR: adjusted odds ratio for age and sex; allele-1: TTAGAG allele of rs17860508; allele-2: GC allele of rs17860508. The significant associations are bolded.

We observed a lower frequency of C-allele and CC-genotype of the IL12B rs3212227 polymorphism among cases with primary BT compared to BrM and controls. The minor allele C of rs3212227 was significantly less frequent in primary BT cases than in controls (16% vs. 26.5% with crude OR = 0.529, 95% CI: 0.3–0.94, p = .029 and adjusted OR = 0.442; 95% CI: 0.23–0.87; p = .018). In addition, the frequency of C-allele was significantly lower among cases with GBM (grade IV gliomas) compared to the controls (14% vs. 26.5%; crude OR = 0.463; 95% CI: 0.230–0.932; p = .031; adjusted OR = 0.486; 95% CI: 0.228–1.036; p = .062). The minor C allele of rs3212227 was associated with decreased risk of primary BT under the dominant model (30% vs. 44.2%; crude OR = 0.542; 95% CI: 0.28–1.04; p = .067; adjusted OR = 0.445; 95% CI: 0.200–0.989; p = .048). There was no significant association between rs3212227 and BrM risk under all analysed genetic models.

The allele-2 of rs17860508 was associated with an increased risk of primary BT as well as of BrM under the codominant, recessive and allelic models (p < .05). The significance was lost after adjustment for age and sex. The distribution of the rs17860508 IL12B polymorphism was approximately similar between cases with primary and metastatic brain tumours.

Serum levels of IL-12p40 and IL-23 in the BT cases

The serum levels of IL-12p40 and IL-23 were significantly higher in all cases with BT than the controls (243.2 ± 355.4 pg/mL vs. 94.8 ± 67.3 pg/mL; p = .00021, and 27.3 ± 38.4 pg/mL vs. 16.8 ± 14.8 pg/mL; p = .04, respectively).

IL-12p40 and IL-23 were in approximately equal concentration in sera obtained by cases with primary BT and BrM as the mean value (283 ± 395.4 pg/mL vs. 226.1 ± 336.9 pg/mL; p = .44 and 24.8 ± 36.1 pg/mL vs. 24.8 ± 38.8 pg/mL; p = .92, respectively).

In addition, as shown in Figure 1, a significantly higher IL-12p40 serum level was detected in men with primary BT than in women (447.1 ± 444.8 pg/mL vs. 130.3 ± 259.1 pg/mL; p = .0007). However, there were no significant differences between men and women with BrM and controls. Interestingly, the IL-23 serum levels were elevated in healthy women in comparison to men (20.57 ± 16.9 pg/mL vs. 10.9 ± 7.7 pg/mL; p = .013).

Figure 1. Serum levels of IL-12p40 (A) and IL-23 (B) in cases with primary brain tumours (BT), brain metastasis (BrM) and controls in relation to sex. The data are mean values ± standard error.

Serum levels of IL-12p40 and IL-23 in association with genotypes of IL12B rs3212227 polymorphism

Concerning the observed sex-specific differences, further serological analyses were made separately for men and women. The serum levels of IL-12p40 of BT cases and controls in relation to IL12B rs3212227 polymorphism and sex are presented in Figure 2.

Figure 2. Serum levels of IL-12p40 in cases with glioblastoma multiforme (GBM), primary brain tumours (BT), brain metastasis (BrM) and controls in association with sex and genotypes of the IL12B rs3212227 polymorphism.

We observed the sex-specific effect of IL12B rs3212227 polymorphism on serum IL-12p40 levels in cases with primary BT contrary to BrM. Among cases with primary BT, the carriership of the variant allele-C was associated with higher IL-12p40 in women, while the same allele was associated with lower IL-12p40 in men. Women with primary BT and the AC + CC genotypes showed higher IL-12p40 than the AA-genotype carriers (301.5 ± 458.7 pg/mL vs. 58.9 ± 73 pg/mL; p = .024), mainly due to cases with GBM (AC + CC vs. AA genotype: 301.5 ± 458.7 pg/mL vs. 36.1 ± 26 pg/mL; p = .062). On the contrary, women with BrM with different genotypes showed similar IL-12p40 levels (AC + CC vs. AA genotype: 323.5 ± 368.5 vs. 244.2 ± 333 pg/mL; p = .46). Men with primary BT, carriers of the AA genotype showed higher IL-12p40 than those with AC + CC-genotypes (703 ± 480.6 pg/mL vs. 242.9 ± 230.2 pg/mL; p = .054). A similar tendency was detected in men with GBM (AA vs. AC + CC genotype: 621.2 ± 510.1 pg/mL vs. 242.9 ± 230.2 pg/mL; p = .067), in contrast to men with BrM (AA vs. AC + CC genotype: 239.5 ± 362.5 pg/mL vs. 135.9 ± 303.5 pg/mL; p = .21).

We also found substantially higher IL-12p40 serum levels in male carriers of the AA genotype with primary BT than in women with the same genotype (703 ± 480.6 pg/mL vs. 58.9 ± 73 pg/mL; p = .009). On the opposite, among the cases with BrM, women carriers of the AC + CC genotypes showed higher IL-12p40 levels than men with the same genotypes (323.5 ± 368.5 pg/mL vs. 135.9 ± 303.5 pg/mL; p = .031).

Serum levels of IL-23 were similar across the patients’ subgroups considering the IL12B rs3212227 polymorphism. However, a higher IL-23 level was observed among male patients with primary BT with the AA-genotype than AC + CC genotypes (22.7 ± 15.7 pg/mL vs. 9.6 ± 9.5 pg/mL; p > .05), contrary to the cases with BrM (14.3 ± 14.1 pg/mL vs. 32.5 ± 52.2 pg/mL; p > .05) without reaching statistical significance.

Serum levels of IL-12p40 and IL-23 in association with genotypes of IL12B rs17860508 polymorphism

The serum levels of IL-12p40 of BT cases and controls in relation to the IL12B rs17860508 polymorphism and sex are presented in Figure 3.

Figure 3. Serum levels of IL-12p40 in cases with glioblastoma multiforme (GBM), primary brain tumours (BT), brain metastasis (BrM) and controls in association with sex and genotypes of the IL12B rs17860508 polymorphism.

Analysis of serum IL-12p40 levels across the genotypes of IL12B rs17860508 showed no statistically significant changes in any subgroup of BT patients. However, we detected significant differences between female and male patients with the same genotype within the same patient subgroup. Women with BrM and carriership of genotype-11 demonstrated a higher IL-12p40 level than men (386.1 ± 411.8 pg/mL vs. 63.4 ± 69.6 pg/mL; p = .033). Men with primary BT and carriership of genotypes-12 + 22 demonstrated a higher IL-12p40 levels than women (593 ± 499.8 pg/mL vs. 143.9 ± 289.5 pg/mL; p = .016). These results confirm the sex-specific differences in brain tumours although should be taken with caution due to the relatively small number of women with primary BT carriers of genotype-11.

The serum levels of IL-23 were similar between cases with genotypes-12 + 22 and those with genotype-11 across all patients subgroups (p > .1)

Discussion

In the present study, we focused on IL12B polymorphisms and investigated their association with the risk of primary and metastatic brain tumours as well as their functional effect on IL-12p40 and IL-23 levels. The C allele of the IL12B rs3212227 polymorphism was associated with a decreased risk of primary BT in contrast to the IL12B rs17860508 polymorphism. In addition, we observed sex-specific differences in the IL-12p40 serum levels in primary BT and identified a differential functional effect of the IL12B rs3212227 polymorphism on cytokine levels by sex.

The minor C allele of the IL12B rs3212227 polymorphism was found to have a protective effect against the risk of primary BT under the dominant and allelic models. The allele-2 of the IL12B rs17860508 polymorphism was correlated with elevated BT risk but just in the models non-stratified by age and sex– a result that indirectly confirms links between age and sex with the development of brain tumours. Previously, the C allele of the IL12B rs3212227 polymorphism was pointed out as a risk factor for glioma [16] in a cohort of Chinese patients with gliomas in different clinical stages. Later, Sima et al. [17] failed to find a significant association between the IL12B rs3212227 polymorphism and the susceptibility to primary brain tumours among Chinese patients. Here, we reported a significant association between the C allele IL12B rs3212227 and decreased risk of primary brain tumours in European patients. Although it is difficult to determine the reason for these contradictory results, some possibilities are different ethnicities, cancer subgroups and other unevaluated factors. The minor allele frequency of rs3212227 for the European population varies from 0.2180 to 0.31, according to the publically available database of Short Genetic Variations repository in the NBCI repository. The observed frequency of the minor allele in our control group (0.265) is close to that and quite different from those reported for Asian subjects within a range of 0.366 to 0.512. Unfortunately, we did not find any case-control study or meta-analysis that has explored the association of the IL12B rs3212227 or rs17860508 polymorphisms with brain tumour risk in the European population. In addition, we should note that the cases with primary BT included in our study were generally with grade IV gliomas (70%), while the cohort of the Zhao et al. [16] study includes cases with all stages of glioma (from stage I to stage IV). Nevertheless, we may hypothesize that the IL12B rs3212227 polymorphism probably influences the primary brain tumour development in contrast to the IL12B rs17860508 polymorphism.

Concerning the effect of the IL12B rs3212227 polymorphism on IL-12p40 expression levels, the carriers of the C allele were associated with decreased IL12B mRNA levels [21] and IL-12p40 secretion after in vitro stimulation of peripheral blood mononuclear cells isolated from healthy volunteers depending on stimuli used [22]. Other studies have reported increased or no effect on cytokine production [23,24]. In our study, we observed a sex-specific functional effect of IL12B rs3212227 on IL-12p40 serum levels among cases with primary BT. The carriership of the variant allele-C was associated with higher serum IL-12p40 in women with primary BT, and lower IL-12p40 in men with primary BT. We observed a similar trend among cases with BrM but without reaching statistical significance. Previously, we have reported sex-specific effects of IL12B gene polymorphisms in relapsing–remitting multiple sclerosis [20], a chronic inflammatory disease affecting the CNS, more common among women than men. Although the molecular mechanisms of this sex-specific effect are unknown and should be confirmed and clarified further, we may speculate that could be due not only to the dynamic interplay among cytokines and hormones but also to the epigenetic regulation of gene expression. МicroRNAs (miRNAs) that have critical roles in the regulation of gene expression usually bind the 3’UTR of the target genes. The rs3212227 polymorphism may potentially alter the miRNAs binding site resulting in varying levels of gene expression. According to the miRTarBase database, IL12B is a target of 7 miRNAs, one of which is hsa-miR-34a-3p playing a key role in regulating the immunological microenvironment of many malignancies, including breast, gastric, lung, glioma, liver, cervix and head and neck cancer [25]. Other studies also point to the sex-specific association between genetic variations and brain cancer. Salnikova et al. [26] suggested sex-specific associations of CYP1A1 and GSTM1 polymorphisms with paediatric brain tumour risk. The mutations in isocitrate dehydrogenases 1 (IDH1), a biomarker of GBM, exhibit sex-specific survival benefits in GBM [27]. More research is required to understand the impact of rs3212227 on brain cancer development and its sex-specific effect. However, we may assume that men, carriers of the C-allele variant associated with lower levels of IL-12p40 could have a decreased risk of development of primary BT.

The sex difference in brain cancer has been a field of research interest for decades. A recent review by Carrano et al. [28] summarizes several differences in GBM between men and women concerning epidemiology, disease phenotype, genetic/molecular factors and outcomes, and underlines the need for sex-stratified studies on brain tumourigenesis. While sex differences in lymphocyte function may contribute to anti-tumour immune response, several previous studies have demonstrated sex dimorphism in the functional activity of the microglia and astrocytes in mice models. The mRNA expression of the pro-inflammatory cytokines IL-1β and IL-6 was significantly higher in microglia isolated from males than in microglia isolated from females [29]. In addition, the male and female cortical astrocytes differ in their response to lipopolysaccharide. The mRNA levels of early pro-inflammatory cytokines, such as IL6, TNFα and IL1β were significantly higher in astrocytes derived from males than in those derived from females [30].

Our result of higher IL-12p40 serum levels in men than women with primary BT is in principal accordance with the above, bearing in mind that IL-12p40 is a subunit shared by IL-12, also known as IL-12p70 (IL12p35/p40) and IL-23 (IL-23p19/p40) that are involved in pro-inflammatory Th1 and Th17 mediated responses, respectively. In addition, serial studies by Kundu et al. [31,32] have identified IL-12p40 as a cytokine with a biological role different from IL-12, IL-23 and homodimer IL-12p40₂. They have shown that monomer IL-12p40 suppressed IL-12Rβ1 internalization, helping certain cancer cells avoid cell death. The authors have demonstrated regression of prostate tumour cells [31] and triple-negative breast cancer cells [32] by monoclonal antibody to p40 monomer, and suggested that the effect is mediated via the IL-12–IFN-γ pathway. In addition, homodimeric IL-12p40₂ was shown as the most potent inducer of IL-16, a leukocyte chemoattractant factor, in mouse and human microglial cells [33]. Together, these findings highlight the importance of the IL-12 family cytokines in neuroimmune dysregulation connected with neurological tumourigenesis and autoimmune disorders.

The present study has some limitations that should be considered. First, the genetic variations in the IL12B gene, including the rs3212227 polymorphism, are only one of the many factors that can contribute to brain tumour risk. More genetic variants, environmental exposure, lifestyle and other factors also play a role in the development of brain tumours. Second, the molecular mechanisms leading to the observed sex-specific effect of IL12B rs3212227 as well as the involvement of IL-12 family cytokines in brain tumourigenesis should be analysed collectively by mRNA expression, protein expression and post-translation modifications assays on local and systemic levels in a sufficiently large sample size in sex-specific stratified models.

Conclusions

On the basis of the obtained results, we may suggest that in Caucasians: (i) IL-12p40 and IL-23 are probably tightly involved in the tumourigenesis of brain tumours, since the serum levels were significantly elevated among cases with primary BT and BrM; (ii) the male gender was associated with higher IL-12p40 among cases with primary BT; (iii) the IL12B rs3212227 polymorphism probably influences the risk of the primary brain tumour in contrast to the IL12B rs17860508 polymorphism and shows a sex-specific functional effect on IL-12p40.

Ethical approval

This study was approved by the institutional ethics committee at Trakia University, Medical Faculty, Bulgaria (decision number 16/19.03.2021).

Consent form

Informed consent was obtained from all subjects enrolled in the study in compliance with the ethical standards of the Helsinki Declaration.

Author contributions

SV, BG and LM were involved in the conception and design; SV, BG, AG, BP, RY and LM planned the experiments; SV, AG, BP, RY and LM performed the analysis; SV, BG, AG, BP, RY and LM were involved in interpretation of the data; BG, AG, BP, and RY were involved in drafting of the paper; SV, and LM revised the manuscript; all authors approved the final version of the paper to be published and are agree to be accountable for all aspects of the work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available upon reasonable request from the corresponding author, LM. The data are not publicly available due to their containing information that could compromise the privacy of research participants.

Additional information

Funding

This work was supported by the Medical Faculty, Trakia University under Grant no. NIP9/2021 and by the Bulgarian Ministry of Education and Science (MES) in the frames of Bulgarian National Recovery and Resilience Plan, Component ‘Innovative Bulgaria’, the Project No.BG-RRP-2.004-0006-C02 ‘Development of research and innovation at Trakia University in service of health and sustainable well-being’.